Remind Me Who I Am

Abstract

After failure, individuals frequently turn to others for support. The current research examined the process through which individuals utilize interpersonal relationships to stabilize threatened self-views. We may seek support to reassure us with warmth and acceptance after a self-threat, or to provide support for threatened self-knowledge. We proposed that although both types of support are likely to repair the affective consequences of a self-threat, only interacting with others who can provide evidence from the individuals’ past that reconfirms a threatened self-aspect would help stabilize the self-concept. Two studies demonstrated that, for individuals who have suffered a self-threat, receiving specific evidentiary support for the threatened self-aspect was more effective at restoring confidence in both the specific self-aspect and at recovering self-concept clarity than was receiving emotional support, whether the interaction was imagined (Study 1), or offered in person (Study 2) after the threat.

Keywords

self-threat self-concept clarity social interactions

“A friend is someone who knows the song in your heart and can sing it back to you when you have forgotten the words.”

—Albert Camus

How do individuals restore their sense of self after a threat? Various failures can question central aspects of self-knowledge, resulting in confusion and distress. Prior research has demonstrated that when individuals attempt to repair self-relevant threats on their own, they frequently invoke indirect repair strategies, focusing on affirming parts of the self-concept that have not been threatened to affirm their overall sense of self as valuable (e.g., Aronson, Blanton, & Cooper, 1995). The current investigation examined the possibility that when core aspects of the self have been threatened, individuals may also turn to others who know them well to remind them of who they are. Specifically, we proposed that individuals scaffold their self-knowledge by seeking direct evidence from others that reconfirms the aspect of the self that has been threatened. In other words, when a threat calls a cherished self-aspect into question, individuals may turn to others to confirm and stabilize the self-concept.

The Self-Concept: Content and Consistency

The self-concept, defined as individuals’ sense of who and what they are (Baumeister, 2010; James, 1890), serves as a cognitive theory that individuals hold about themselves (Epstein, 1973). Within this theory, individuals’ selves are hierarchically organized, possessing multiple specific self-aspects (e.g., “I am an artist”) created and supported by their individual and social experiences (e.g., Markus, 1977; Markus & Wurf, 1987; McConnell, 2011). Individuals gain “data” supporting their self-aspects through direct observation of their own behaviors and access to their own memories (e.g., Bem, 1972) as well as from the their social interactions and relationships (e.g., Aron, 2003; Goffman, 1959; Mead, 1934; Tice & Baumeister, 2001). Thus, someone who possessed the self-aspect “artist” might have that aspect supported in multiple ways—by stored self-knowledge such as memories of art school, by reflected self-views such as compliments from friends on their artistic talent, and by tangible evidence of their artistic nature in the form of awards or paintings shown in galleries.

Individuals’ self-aspects are organized into an overarching “theory” of self, and the organization of the self-concept as consistent across time and attributes is essential to individuals’ well-being (e.g., McConnell, 2011). Self-concept clarity is a metacognitive construct that assesses the extent to which individuals perceive their self “theory” as clear and coherent. Self-concept clarity encompasses “the extent to which the contents of an individual’s self-concept (e.g., perceived personal attributes) are clearly and confidently defined, internally consistent, and temporally stable” (Campbell et al., 1996, p. 141). High self-concept clarity is associated with a variety of positive personality characteristics and psychological adjustment (e.g., Bigler, Neimeyer, & Brown, 2001; Campbell, Assanand, & Paula, 2003; Ritchie, Sedikides, Wildschut, Arndt, & Gidron, 2011).¹ Although most often studied as an individual difference, situations that threaten a particular aspect of the self can also reduce individuals’ momentary experiences of self-concept clarity (Nezlek & Plesko, 2001; Slotter, Gardner, & Finkel, 2010).

The Threatened Self: Indirect Repair Strategies

Individuals’ experiences constantly test the validity and coherence of their self-concept. Specifically, novel information can either be consistent with the self-concept, thus confirming the individuals’ existing theories of who they are, or can be inconsistent with the self-concept, thus threatening individuals’ existing theories of who they are. Threats to the self-concept whether from threats to the content of the self (e.g., Festinger, 1957; Steele, 1988; Swann, 1983), or social threats more generally (e.g., Baumeister & Leary, 1995; Pickett & Gardner, 2005; Slotter et al., 2010), produce emotional distress for individuals.

Individuals seem loathe to change their firmly held, existing self-theories in the face of inconsistent information (cognitive dissonance is an exception; Festinger, 1957). Instead, they utilize a wide range of strategies to defend their existing self-concept from threat. Some of these are fundamentally intrapersonal, occurring within the mind of the individual. For example, individuals self-enhance by selectively attending to and remembering positive versus negative information about the self and defend against incoming negative self-relevant information by discrediting it (for reviews, see Alicke & Sedikides, 2009; Sedikides & Gregg, 2008). Individuals may also affirm other central and important self-aspects unrelated to the threat to maintain an overall positive self-view (e.g., Steele, 1988). An interesting facet of intrapersonal strategies for maintaining the self is that individuals often seek repair in an indirect manner by affirming themselves as positive and worthwhile individuals at the general level of the overarching self-concept rather than affirming the specific self-aspect that was threatened (e.g., Crocker, Niiya, & Mischkowski, 2008; Steele, 1988). Indeed, when individuals do affirm a specific self-aspect after a threat to their self-concept, it is often an aspect other than the one that was threatened (e.g., Aronson et al., 1995; Brown & Smart, 1991; Greenberg & Pyszczynski, 1985; see also Tesser, 2000). Researchers have proposed that directly affirming a threatened self-aspect actually draws attention to the threatening information, exacerbating the threat (e.g., Aronson et al., 1995).

The Threatened Self: Direct, Evidence-Based Repair

Two notable exceptions to these indirect repair strategies are self-verification (e.g., Swann, 1983) and self-completion (e.g., Wicklund & Gollwitzer, 1982). When seeking self-verification, individuals actively choose interaction partners and social situations that help them to verify their existing self-beliefs, choosing to interact with others who view them as they view themselves. When engaging in self-completion, individuals directly demonstrate threatened self-aspects in social settings to validate a new or threatened self-aspect. Research from these frameworks showcase the strategies individuals can utilize to directly repair a firmly held self-view. For example, Swann and Hill (1982) demonstrated that individuals faced with inconsistent social feedback (e.g., that an interaction partner saw them as submissive when they viewed themselves as dominant) would only change their self-views when they were denied the opportunity to subsequently display their “true selves” to the interaction partner. Those who were given the opportunity to behave in a more self-confirmatory manner to their interaction partner did not change their self-views. Individuals engage in these behavioral displays even toward novel social others. In a study conducted by Gollwitzer and Wicklund (1985), individuals who were told their personalities were inconsistent with a cherished role (e.g., journalist, musician) claimed extreme competence in these domains to a new interaction partner. Thus, one way individuals address self-threats directly is by demonstrating the possession of the aspect under threat, or in the purview of Epstein (1973) by creating “evidence.”

In each of these prior cases, individuals appear to use the social setting merely as an arena within which they can display behavioral evidence of the threatened self-aspect. This fact was most elegantly demonstrated by Swann and Hill (1982), who directly assessed whether participants believed their interaction partners had revised their views of them, and showed that participants did not; the behavioral evidence they created for themselves was sufficient to protect their self-views from change, regardless of whether they changed their interaction partner’s beliefs. Of course, the beliefs and memories of individuals that are held by social partners can themselves serve as repositories of evidence. For example, a mother is likely to remember multiple events in which her son behaved in alignment with his self-views. Should the son be given feedback that, for example, he was submissive when he thinks of himself as dominant, Mom could then reassure him by recounting the many instances in which she witnessed his dominant behavior. To the extent she also generally views him as dominant, she may even be motivated to help him defend this self-view; research has demonstrated that individuals appeared as motivated to defend their impressions of their spouses from impression inconsistent feedback as their spouses were to defend their self-views from the same inconsistent feedback (De La Ronde & Swann, 1998). Thus, this strategy, turning to others to support us with direct evidence supporting our self-views, appears to co-exist with other direct repair strategies such as demonstrating such evidence ourselves.

The current work explores the extent to which the repositories of evidence held in the minds of others can be successfully utilized by individuals to protect a threatened self-view and thus stabilize an unstable self. We sought to extend the work showing that we prefer close others who see as we see ourselves to examine the precise content of our self-verifying interactions with these people. Specifically, we wished to investigate whether particular types of supportive interactions would be particularly effective in aiding individuals in shoring up their de-stabilized self-concept after a threat to a cherished aspect of who they are. Drawing from the previously discussed literature on the self-concept as a theory (Epstein, 1973), receiving evidence supporting the threatened self-aspect from others might be key for helping individuals to overcome a threat both to that specific self-aspect and to restore clarity of the self-concept more generally. We termed this form of support evidentiary support.

Evidentiary support occurs when social others provide individuals with data that directly address and reconfirm a threatened aspect of their self-concept. A variety of social others should be able to provide evidentiary support as it is fundamentally about providing data to individuals about the threatened aspect of their self-concepts. The data that others provide should be drawn from the social other’s knowledge of and experiences with the specific individual, aimed at reminding the individual that he or she does, in fact, possess the self-aspect that has been called into question. For example, imagine an individual thinks about himself or herself as artistic yet experiences a threat to this self-aspect. A friend could provide evidentiary support by reminding the individual of past social acknowledgment of his or her “artistic” self-aspect, such as past successful exhibitions, or prior compliments received, or past positive reviews of his or her work. Evidentiary support would also encompass reminders of specific behavioral instances that could confirm his or her threatened “artistic” self-aspect, such as recalling the individual’s spending hours in the studio producing beautiful paintings, or being selected for art camp as a child, or artistic talents displayed through other means—such as exquisite aesthetic taste or decorating advice that showcased an artistic flair.

This interpersonally provided evidence can then be used to scaffold and stabilize the threatened aspect of the self-concept. To the extent the threatened aspect is one central to their sense of self, then evidence supporting that aspect should stabilize both the specific self-view under threat and the individuals’ metacognitive view of themselves as clear and consistent. A burgeoning body of literature demonstrates that this self-concept clarity can be situationally reduced (Nezlek & Plesko, 2001; Slotter et al., 2010), and it is possible that evidentiary support may be helpful in restoring it. Evidentiary support thus provides a complementary strategy to the direct self-concept repair strategies previously demonstrated within the self-verification and self-completion literatures, by focusing on the memories and behaviors of the threatened individual’s relationship partners rather than those of the individual herself or himself. It is clear that when strongly held self-aspects are threatened, individuals’ self-views can be protected when given a chance to display evidence of those qualities (e.g., Gollwitzer & Wicklund, 1985; Swann & Hill, 1982). It is also likely that individuals may turn to others for support. Indeed, individuals facing self-inconsistent feedback can be protected from self-change by interacting with close others who see them as they see themselves (Swann & Predmore, 1985). Moreover, close others are often themselves invested in the stability of the impressions they hold regarding a partner—for example, spouses faced with feedback threatening their impressions of their spouse (e.g., that a wife performed poorly on a social skills inventory, when the husband believes her to be socially skilled) appear to struggle with the feedback, actively disagreeing with it and defending their own impressions as intensely as the individual whose own self-beliefs were threatened (De La Ronde & Swann, 1998). From this work it is clear that close others are not only capable of defending our self-views but also often both likely and motivated to do so—and couples for whom self-views and spousal impressions can be maintained in harmony experience greater intimacy (De La Ronde & Swann, 1998). The exact support mechanism and/or the exact interaction content, through which close others can help best stabilize self-views, however, remains unclear. We propose that the type of social interaction most protective for self-views will be one characterized by evidentiary support, when others are reminding the threatened individual of her or his past behavior and characteristics, perhaps even in ways that she or he herself or himself cannot.

Evidentiary Support Within the Context of the Social Support Literature

Several taxonomies of types of supportive social interactions distinguish support that is focused on exchanges of emotional warmth, empathy, and care from support that is focused on rational, task-based advice and guidance (e.g., Cohen & McKay, 1984; Harlow & Cantor, 1995; Taylor, 2007), roughly dividing types of types of supportive interactions into emotional versus task-based functions (see House, 1981, for a further breakdown of support taxonomies). Importantly, the benefits of support for the individual appear to vary by the situation that prompted the need for support in the first place (e.g., Cresci, 2001; Cutrona & Russell, 1990; Malecki & Demeray, 2003; Taylor, 2007), with appropriately formed support efforts promoting the greatest benefit for well-being. For example, Cutrona (1990) suggested that emotional support can reduce individuals’ emotional distress in the face of threat but may predict a withdrawal of effort in overcoming the threat. Generally speaking, research has suggested that support that is task-focused is better suited to aiding individuals in overcoming failures during task pursuit than emotional support (e.g., Cutrona & Russell, 1990). Thus, when individuals are faced with a threat to a cherished self-aspect, receiving support from others that provides them with the evidence necessary to refute that threat, particularly when they are unable to draw on evidence they have gathered themselves, should be the most beneficial in helping them stabilize their now uncertain sense of self.

Research Overview and Hypotheses

The current research examined whether, as the self-concept is a fundamentally evidence-driven knowledge structure, receiving evidence from others that directly supports a threatened aspect of the self would stabilize the threatened self-aspect and thus, allow recovery of self-concept clarity. We termed this form of interaction evidentiary support and contrasted it with a common form of social support, emotional warmth that is characterized by warm, empathic interactions that focus on the individual as worthy of affection. Emotional warmth has been shown to assist individuals in coping with a wide variety of stressors (e.g., Cresci, 2001; Crocker et al., 2008; Cutrona, 1990; Leary & Baumeister, 2000; see also Taylor, 2007). However, despite the well-established positive role of warm, supportive interactions in well-being, we did not believe receiving emotional warmth would be sufficient to repair the self, with regard to content and consistency, after a threat. Although emotional warmth may remind individuals of their social connections and value, much like self-affirmation, thus making them feel better after the threat, these types of interactions, also like self-affirmation, do not offer direct confirmation of the specific aspect of the self that has been threatened and thus should not represent the optimal form of social support for this particular threat (e.g., Crocker et al., 2008; Cutrona, 1990). We hypothesized that receiving evidentiary support for a threatened self-aspect should confirm both the specific self-aspect in question and bolster individuals’ overall self-concept clarity (Studies 1 and 2). We also explored the potential independence of self-concept repair and mood repair, hypothesizing that receiving any type of support from a friend would be sufficient to repair the emotional consequences of the self-threat, but that only evidentiary support would repair the threatened self-aspect and thus maintain self-concept clarity (Study 2).

Study 1

Study 1 tested our hypothesis that receiving evidentiary support would be more beneficial than emotional warmth for repairing individuals’ self-concepts after threat. Participants all failed at a task that either did or did not threaten a self-aspect (self-as-doctor) that was either important to the participant or not. They then imagined interacting with either a friend who could offer evidentiary support, a friend who could offer emotional support, or were given a control task. Specifically, we examined changes in individuals’ (a) certainty that they possessed the specific threatened self-concept and (b) overall self-concept clarity, as a function of the type of social interactions they imagined having after a threat to their self-concept. According to hierarchical models of self-organization, higher order perceptions of the self are supported through the stability and consistency of lower level self-aspects (e.g., McConnell, 2011). Thus, we also examined whether any relationship found between evidentiary support and increased self-concept clarity would be mediated through changes in individuals’ certainty that they possessed the specific threatened self-aspect. We predicted that only individuals who hoped to be a doctor in the future and who imagined evidentiary support after a self-threat would remain confident in their future self as a doctor, and that this would result in maintaining higher self-concept clarity after the threat.

Method

Participants

One hundred and eighteen undergraduates (69 women) took part in the current study in partial fulfillment of the requirements for a psychology course. Participants were 18.57 years old (SD = 0.84) on average.

Procedure

During a pre-testing session, participants reported on whether they intended to go to medical school after receiving their bachelor’s degree, their current or intended major, and their career plans in the future. Sixty-seven of the participants in the current study self-reported that they were pre-medical majors, intended to go to medical school and to become a doctor. In contrast, 51 of the participants self-reported that they were humanities or social science majors who did not intend to go to medical school or become doctors. In addition, as described below, participants rated themselves on a variety of self-aspects as part of the pre-testing session. One of the self-aspects embedded in this list was the aspect of “doctor.” To be included in the study, participants in the pre-medical major group had to rate themselves above a 1 (the lowest rating on the scale) on the “doctor” item; all of them did. To be included in the study, participants in the non-pre-medical major group had to rate themselves a 1 on the “doctor” item. Ten pre-screened participants who would have fit in the non-pre-medical major group rated themselves higher on the “doctor” item and were excluded from the study prior to data collection, resulting in the 118 participants reported above.

Participants came to the lab a minimum of 2 weeks later and completed a series of tasks ostensibly to assess whether or not they are capable of being a successful doctor. First, participants completed five questions taken from practice materials for the Medical College Admissions Test (MCAT). Second, they read two synopses about patients who were experiencing ambiguous symptoms and had to select which follow-up questions they would ask each patient first if they were the diagnosing physician. Both questions were plausible answers in general—participants had the option of asking about family history of disease or about the patient’s medications—but the experimenter told participants that one of the two questions was the correct one to ask first. Importantly, the experimenter framed the two diagnostic problems as an assessment of participants’ diagnostic intuition, which was described as an innate and largely unalterable ability that was important in making medical decisions. All participants then received feedback that they had incorrectly answered four out of the five MCAT questions and they had selected the incorrect follow-up questions to ask both hypothetical patients, leading to misdiagnosis and delayed treatment that could potentially cause the patients suffering and life-threatening medical complications. Thus, they exhibited both low levels of medical knowledge and poor diagnostic intuitions and were unlikely to be successful medical doctors in the future.

Participants then were randomly assigned to one of three imagined support conditions. All participants were asked to imagine talking to someone in their life about their poor performance. In the evidentiary support condition, they were asked to imagine and write about talking to someone who knew them well and could remind them of who they are if they felt unsure, completing three prompts answering what the friend might say to them about why they would make good doctors. In the emotional warmth condition, participants were asked to imagine and write about talking to someone in their life who was emotionally warm and someone they could count on to make them feel happy, completing three prompts regarding why the friend values them as a person . . . In the control condition, participants were asked to recall their trip to the laboratory that day and complete three prompts regarding their experiences on the trip. They then completed a series of survey measures, including a measure of their perceptions of themselves as doctors, followed by a measure of their self-concept clarity.

To obtain adequate power, 120 participants were recruited: 2 never came in. At the conclusion of the study, we completed a suspicion check asking participants whether they believed the feedback they received regarding their ability to be a doctor. All participants reported believing the feedback resulting in the 118 participants reported above.

Measures

“Doctor” self-ratings

Participants rated how characteristic a series of 15 mildly positively valenced self-aspects, taken from Anderson’s (1968) work on personality traits, were of them (1 = not at all, 7 = extremely). Central to the current study, the self-aspect of “doctor” was embedded in the list at the pre-testing session (M = 3.76, SD = 2.53) and the list at the end of the experimental session (M = 3.64, SD = 2.14).

Self-concept clarity

Participants completed a well-validated 12-item measure of self-concept clarity at both the pre-testing session (α = .91, M = 4.10, SD = 1.29) and after the experimental manipulation (α = .92, M = 4.26, SD = 1.34; 1 = strongly disagree, 7 = strongly agree; Campbell et al., 1996).

Results and Discussion

We first tested our central hypothesis that receiving evidentiary support from others would better be able to repair the threatened self-aspect and stabilize the self-concept than imagining receiving emotional warmth or a control scenario. In the current study, we expected that this would only be true if the threat was self-concept relevant. Thus, the benefits of evidentiary support after a failure experience should only extend to individuals for whom the failure threatens an important self-aspect: participants in the pre-medical majors group. We conducted two 2 Major Group (pre-medical, non-premedical) × 3 Support Condition (evidentiary, emotional warmth, distraction) ANCOVAs examining participants’ post-manipulation (a) “doctor” self-ratings and (b) self-concept clarity controlling for their pre-testing rating of the relevant construct.^2,3

“Doctor” self-ratings

As predicted, a main effect of group emerged, such that participants in the pre-medical majors group endorsed “doctor” as more characteristic of them (M = 5.61, SD = 1.48) than participants in the non-pre-medical majors group (M = 1.65, SD = 0.82), F(6, 111) = 19.57, p < .001, $η_{p}^{2}$ = .15, 95% confidence interval (CI) = [0.63, 2.09]. A main effect of support condition also emerged, F(6, 111) = 4.04, p = .02, $η_{p}^{2}$ = .07, 95% CI = [−0.51, −0.03]. However, these main effects were qualified by an interaction between group and support condition (Figure 1), F(6, 111) = 10.31, p < .001, $η_{p}^{2}$ = .16, 95% CI = [−0.71, −0.17]. No effect of pre-test “doctor” self-ratings emerged, F(6, 111) = 0.30, p = .58, $η_{p}^{2}$ = .002, 95% CI = [−0.28, 0.18].

Figure 1.

Study 1: Perception of the self-aspect “doctor” as characteristic of the self-concept as a function of assigned support condition.

We next examined the simple effect of support condition within pre-medical versus non-pre-medical majors. As predicted, support condition predicted “doctor” self-ratings among participants in the pre-medical majors group, F(6, 111) = 16.37, p < .001, but not among participants in the non-pre-medical majors group, F(6, 111) = 0.67, p = .52. Planned contrasts of the effect of different levels of support condition within the pre-medical majors group revealed that participants perceived the self-aspect “doctor” to be more characteristic of their self-concept after imagining receiving evidentiary support (M = 6.22, SD = 0.80) compared with after imagining receiving either emotional warmth (M = 4.36, SD = 1.76), t(111) = 5.48, p < .001, or imagining their walk to the lab (M = 4.86, SD = 1.08), t(111) = 4.10, p < .001. No difference emerged between the emotional warmth and control conditions, t(111) = 1.42, p = .16.

Self-concept clarity

No significant main effect of group emerged, F(6, 111) = 1.14, p = .28, $η_{p}^{2}$ = .01, 95% CI = [−0.26, 0.10]; however, a main effect of support condition did emerge, F(6, 111) = 5.33, p < .01, $η_{p}^{2}$ = .08, 95% CI = [−0.37, −0.11]. This main effect was qualified by a significant interaction between group and support conditions (Figure 2), F(6, 111) = 3.32, p < .05, $η_{p}^{2}$ = .06, 95% CI = [−0.22, −0.07]. Pre-testing self-concept clarity was a significant predictor of post-manipulation self-concept clarity, F(6, 111) = 8.54, p < .001, $η_{p}^{2}$ = .44, 95% CI = [0.53, −0.83].

Figure 2.

Study 1: Self-concept clarity as a function of assigned support condition.

We next examined the simple effect of support condition within pre-medical versus non-pre-medical majors. As predicted, support condition predicted self-concept clarity among participants in the pre-medical majors group, F(6, 111) = 9.76, p < .001, but not among participants in the non-pre-medical majors group, F(6, 111) = 0.35, p = .70. Planned contrasts of the effect of different levels of support condition within the pre-medical majors group revealed that participants in the pre-medical majors group reported experiencing greater self-concept clarity receiving evidentiary support (M = 4.89, SD = 1.03) compared with after imagining receiving either emotional warmth (M = 3.73, SD = 1.49), t(111) = 4.41, p < .001, or imagining a control scenario (M = 4.17, SD = 1.07), t(111) = 2.38, p < .05. A marginal difference emerged between the emotional warmth and control conditions such that self-concept clarity was slightly higher in the control condition, t(111) = 1.08, p = .06.⁴

Mediation

Next, we tested whether participants’ endorsement of the “doctor” self-aspect mediated the relationship between support condition and self-concept clarity. As participants in the non-pre-medical majors group generally did not perceive “doctor” to be characteristic of them at either the pre-testing or experimental session, we only examined participants who were in the pre-medical majors group. We coded support condition (evidentiary = 1, emotional = −1, distraction = 0) and standardized all other variables prior to analyses (M = 0, SD = 1). We controlled for participants’ pre-testing endorsement of “doctor” and self-concept clarity in all relevant analyses.

We used the bootstrapping approach advocated by Preacher and Hayes (2004) based on 5,000 resamples. As established by our earlier analyses, our independent variable, support condition, predicted our dependent variable, self-concept clarity, β = .21, t(64) = 2.13, p < .05, and our independent variable, support condition, predicted our mediator, endorsement of “doctor,” β = .38, t(64) = 3.31, p < .01. When entered simultaneously into the regression with support condition, greater self-ratings of “doctor” predicted greater self-concept clarity (Figure 3), β = .24, t(61) = 2.26, p < .05. Assigned support condition did not remain a significant predictor of self-concept clarity in this analysis, β = .12, t(61) = 1.17, p = .25, and revealed a total indirect effect of condition with a 95% bias corrected and accelerated (BCa) bootstrap CI of 0.04 to 0.21. Zero falls outside of this CI, indicating significant mediation at the p < .05 level.

Figure 3.

Study 1: Perception of the self-aspect “doctor” as characteristic of the self-concept mediates the relationship between assigned support condition and self-concept clarity for individuals who want to be medical doctors.

The results of Study 1 demonstrated that imagining receiving evidentiary support, but not emotional warmth or a control scenario, increased individuals’ certainty that they possessed the specific self-aspect that had been threatened, and increased individuals’ self-concept clarity, but only for individuals who identified with becoming a medical doctor. Thus, evidentiary support aided individuals in confirming their threatened self-concept but only when the threat was to an important self-aspect (e.g., Tesser, 2000). In addition, Study 1 found evidence for a mediational pathway such that participants’ increased certainty that they possessed the specific threatened self-aspect significantly accounted for the relationship between imagining receiving evidentiary support after a self-concept threat and increased self-concept clarity.

Study 2

Study 2 expanded on the previous results in three important ways. First, Study 2 examined the influence of evidentiary support versus emotional warmth in live interactions between participants and a close other. Second, Study 2 incorporated measures of mood, so that mood repair could be tested in addition to self-concept repair. Finally, participants’ interactions were audio-taped so that objective coders could rate the participants’ levels of self-certainty with regard to the specific threatened aspect, as well as the type of support provided by the close other.

Method

Participants

Ninety undergraduates (53 women) took part in the current study in partial fulfillment of the requirements for a psychology course. Participants were 18.63 years old (SD = 0.78) on average.

Procedure

During a pre-testing session that took place a minimum of 2 weeks prior to any laboratory sessions, participants reported on whether they intended to go to medical school after receiving their bachelor’s degree and whether they wanted to be a medical doctor in the future. Only participants who responded affirmatively to both questions were selected to participate in the current study.

Once selected, participants completed all portions of the study in a single experimental session. Before their arrival to the laboratory, participants received instructions to bring their closest, on campus, same-sex friend with them to the study. On their arrival to the laboratory, the participants were taken to a separate room to complete the same series of medical tasks from Study 1, followed by self-concept threatening feedback from Study 1. They then completed a measure of positive and negative emotions and the Self-Concept Clarity Scale used in Study 1.

While the participants completed the self-threat and survey measures, the experimenter took the friends who had accompanied the participants into a separate room and gave them instructions regarding the support they were to provide the participants. Specifically, the experimenter told the friends that the participant was selected to participate in this study because he or she reported that he or she wanted to become a doctor on the pre-testing measure. They were told that the participant was in another room completing a variety of tasks, such as MCAT problems and diagnostic problems, which would assess the likelihood that the participant would be a good doctor. They were also told that they would be having a 5-min interaction with the participant in a few minutes and that they should ask the participant about his or her performance on the tasks.

The experimenter then provided the friend with one of the three different sets of instructions that corresponded to randomly assigned support conditions. In the evidentiary support condition, the experimenter told the friends that, if the participant had not done well on the medical tasks, they should remind the participant about all of the evidence, or data, which exists that they know about that indicates that the participant would, in fact, make a good doctor. The friends were told to reference the participant’s grades, course selection, or diagnostic abilities (i.e., if the participant told their friend to go to the doctor or was able to suggest a remedy that made their friend feel better when they were ill.). The friends were asked to focus on being matter-of-fact, focus on the evidence that they had about the participant, and avoid being overly emotional. In the emotional warmth condition, the experimenter told the friends that if the participant had not done well on the medical tasks, they should remind the participant that they care about him or her and value him or her as a friend. The friends were told to reference that the participant was a good friend and lovable person who they valued. The friends were asked to focus on expressing empathy, emotional warmth, and understanding, and avoid directly mentioning any specifics relevant to being a doctor, such as grades or coursework. In the distraction condition, the experimenter told the friends that if the participant had not done well on the medical tasks, they should simply distract the participant from the upsetting failure. The friends were told to talk about potential plans for the upcoming weekend with the participant and to change the topic if the participant brought up his or her performance on the medical tasks.

After the experimenter provided the friends with the appropriate instructions, and the participants completed the medical tasks and survey measures, participants returned to the room where their friends were seated and engaged in a 5-min interaction with their friends—who were told to behave in accordance with the instructions they received—which was audio recorded with the participants’ and friends’ knowledge. Prior to any analysis, the audio tapes were reviewed for compliance with the study procedures and to probe for suspicion.

After engaging in their 5-min interaction, the participants returned to the room where they had completed the medical tasks, and completed the mood and self-concept clarity scales a second time. After data collection was completed, the audio-taped interactions were coded for the extent of evidentiary support and emotional warmth provided by the friends, as well as participants’ expressed certainty in themselves as future doctors.

To obtain adequate power, 100 participants were initially recruited, 2 never came in, and 8 participants and friends did not complete the interaction as requested or expressed suspicion regarding the feedback they had received. These 8 participants were discarded from the data set prior to analysis, resulting in the 90 participants reported above.

Self-Report Measures

Mood

Participants completed a measure of their current mood, the Positive and Negative Affect Schedule (PANAS; Watson, Clark, & Tellegen, 1988). They rated a series of 20 affective terms, 10 positive and 10 negative, for how much they were feeling each at the present moment (1 = not very much, 5 = very much). They rated both sets of items after the threat (Positive items: α = .90; M = 2.84; SD = 0.78; Negative items: α = .91; M = 1.53; SD = 0.47), then again after their interaction with their friend (Positive items: α = .90; M = 2.98; SD = 0.89; Negative items: α = .90; M = 1.30; SD = 0.43).

Self-concept clarity

Participants completed the same Self-Concept Clarity Scale used in Study 1 both after the threat (α = .88; M = 4.73; SD = 1.08) and after the interaction with their friend (α = .85; M = 4.80; SD = 1.00; Campbell et al., 1996).

Coding procedures and measures

Two objective coders, who were blind to participant condition, rated the recorded linguistic content of the interactions that the participants had with their friends for a variety of relevant constructs. Prior to coding, the audio recordings of participants’ interactions with their friend were transcribed and coders rated both the friends’ and the participants’ statements across the interaction on a variety of constructs.

Coded evidentiary support and emotional warmth

Coders rated the extent to which the friend provided participants with informational and evidence-focused support (coded evidentiary support; Intra-Class Correlation (ICC) = .88, M = 1.44, SD = 1.74) and emotional warmth or empathy (coded emotional warmth; ICC = .86; M = 1.29, SD = 1.45; all codes 0 = no support, 6 = extreme support). For example, one friend, who received a mean coded evidentiary support rating of 4.0 across the two coders made statements such as “You will be a great doctor! You get such good grades in all of your classes. You rocked orgo [organic chemistry].” One friend who received a mean coded emotional warmth rating of 4.5 across the two coders made statements such as “Sweetie, I’m sorry! You know that I’m here if you need a hug.”

Coded “doctor” self-certainty

Coders also rated how certain participants were of possessing the self-aspect “doctor” at the beginning of the interaction (the first minute of talking; ICC = .70; M = −0.15, SD = 1.44) and at the end of the interaction (the last minute of talking; ICC = .71; M = 0.75, SD = 1.37; all codes −3 = not at all certain, 3 = extremely certain). All 90 participants had usable data for these two time periods. For example, one participant, who received a mean coded “end of interaction” rating of −1.5 across the two coders made statements such as “My dad is a surgeon, you know. So, if I can’t make it as a doctor, I should just quit now . . . ” In contrast, a different participant, who received a mean coded “end of interaction” rating of 1.5 across the two coders made statements such as “I’m still applying to medical school. Getting a couple of questions wrong doesn’t mean I can’t be a doctor.”

Results and Discussion

We first conducted a manipulation check in which we conducted two 3 Support Condition (evidentiary, emotional warmth, distraction) ANOVAs predicting coded evidentiary support or emotional warmth, respectively. Next, to test our central hypothesis that receiving evidentiary support from others would better be able to repair the threatened self-aspect and stabilize the self-concept than receiving emotional warmth or a control scenario, we conducted a series of three 3 Support Condition (evidentiary, emotional warmth, distraction) ANCOVAs, examining participants’ post-manipulation (a) coded “doctor” self-certainty, (b) self-reported self-concept clarity, and (c) positive and negative mood, controlling for the earlier code or rating of the relevant construct.^5,6

Manipulation check

As predicted, when examining coded amounts of evidentiary support, a significant effect of condition emerged, F(2, 88) = 57.80, p < .001, $η_{p}^{2}$ = .57, 95% CI = [0.47, 0.64]. Planned contrasts revealed that participants’ interactions with their friend were characterized by greater verbalized evidentiary support in the evidentiary support condition (M = 3.25, SD = 1.72), compared with the emotional warmth condition (M = 1.20, SD = 1.10), t(88) = 6.50, p < .001, or the distraction condition (M = 0.02, SD = 0.01), t(88) = 10.72, p < .001. Interestingly, the emotional warmth condition was also characterized by greater evidentiary support than the distraction condition, t(88) = 4.45, p < .001, although, still less than the evidentiary support condition. Similarly, when examining coded amounts of emotional warmth, a significant effect of condition emerged, F(2, 88) = 46.65, p < .001, $η_{p}^{2}$ = .52, 95% CI = [0.42, 0.59]. Planned contrasts revealed that participants’ interactions with their friend were characterized by greater verbalized emotional warmth in the emotional warmth condition (M = 2.58, SD = 1.39), compared with the evidentiary support condition (M = 1.13, SD = 1.01), t(88) = 5.44, p < .001, or the distraction condition (M = 0.11, SD = 0.25), t(88) = 9.61, p < .001. The evidentiary support condition was also characterized by greater emotional warmth than the distraction condition, t(88) = 3.79, p < .01, although, still less than the emotional warmth condition. Taken together, our manipulation appeared successful; thus, we analyzed tests of our central hypothesis using assigned support condition as the independent variable.

Coded “doctor” self-certainty by condition

As predicted, when examining coded “doctor” self-certainty, an effect of condition emerged, F(2, 87) = 14.04, p < .001, $η_{p}^{2}$ = .24, 95% CI = [0.14, 0.35]. Beginning of interaction “doctor” self-certainty was a significant predictor of end of interaction “doctor” self-certainty in this analysis, F(2, 87) = 7.03, p < .001, $η_{p}^{2}$ = .06, 95% CI = [0.01, 0.14]. Planned contrasts revealed that coders rated participants as having greater “doctor” self-certainty after receiving evidentiary support (M = 5.15, SD = 0.98) compared with after receiving either emotional warmth (M = 4.38, SD = 1.01), t(87) = 3.08, p < .01, or distraction (M = 4.84, SD = 1.07), t(87) = 5.27, p < .01. No significant difference emerged between the emotional warmth and control conditions, t(87) = 1.54, p = .08.

Self-concept clarity and mood by condition

As predicted, when examining self-concept clarity, an effect of condition emerged, F(2, 87) = 5.69, p < .01, $η_{p}^{2}$ = .07, 95% CI = [0.01, .014]. Pre-interaction self-concept clarity was a significant predictor of post-interaction self-concept clarity in this analysis F(2, 87) = 23.87, p < .001, $η_{p}^{2}$ = .39 95% CI = [0.28, 0.48]. Planned contrasts revealed that participants reported experiencing greater self-concept clarity after receiving evidentiary support (M = 5.15, SD = 0.98) compared with after receiving either emotional warmth (M = 4.38, SD = 1.01), t(87) = 3.14, p < .01, or distraction (M = 4.84, SD = 1.07), t(87) = 2.68, p < .01. No difference emerged between the emotional warmth and control conditions, t(87) = .47, p = .62.⁷ In contrast, no significant effects of condition emerged when predicting participants’ post-interaction positive, F(2, 87) = 0.76, p = .46, $η_{p}^{2}$ = .01, 95% CI = [0.02, 0.04], or negative mood, F(2, 87) = 0.66, p = .52, $η_{p}^{2}$ = .01, 95% CI = [−0.02, 0.05]. In each of these analyses, pre-interaction positive, F(2, 87) = 113.27, p < .001, $η_{p}^{2}$ = .56, 95% CI = [0.46, 0.63], and negative mood, F(2, 87) = 51.91, p < .001, $η_{p}^{2}$ = .37, 95% CI = [0.26, 0.47], did predict post-interaction ratings.

In addition, we ran three 2 (Pre- vs. Post-Interaction self-concept clarity/positive affect/negative affect) repeated-factors ANOVAs to verify that simply interacting with a friend, irrespective of support condition, was insufficient to boost self-concept clarity from post-threat to post-interaction, but would be sufficient to boost mood. As predicted, no main effect of self-concept clarity from pre- to post-interaction emerged, F(1, 87) = 0.001, p = .97, $η_{p}^{2}$ = .01, 95% CI = [−0.02, 0.04]. However, for positive, F(2, 87) = 6.37, p = .01, $η_{p}^{2}$ = .06, 95% CI = [2.77, 3.12], and negative, F(2,87) = 6.21, p < .01, $η_{p}^{2}$ = .38, 95% CI = [1.42, 1.62], affect, a pre- to post-interaction effect emerged such that interacting with a friend boosted positive (M_pre = 2.82, SD = 0.78; M_post = 2.95, SD = 0.89) and reduced negative mood (M_pre = 1.52, SD = 0.47; M_post = 1.29, SD = 0.42). Taken together, these results indicate that different types of supportive interactions differentially influenced participants’ self-views but did not differentially affect participants’ moods.

Mediation

Next, we examined whether participants’ coded certainty of themselves possessing the self-aspect of “doctor” mediated the relationship between support condition and self-concept clarity. We coded support condition (evidentiary = 1, emotional = −1, distraction = 0) and standardized all other variables prior to analyses (M = 0, SD = 1). We controlled for participants’ beginning of the interaction codes of “doctor” self-certainty and pre-interaction self-concept clarity in all relevant analyses.

As in Study 1, we used the approach advocated by Preacher and Hayes (2004) based on 5,000 resamples. As established by our earlier analyses, our independent variable, support condition, predicted our dependent variable, self-concept clarity, β = .44, t(87) = 3.17, p < .01, and our independent variable, support condition, predicted our mediator, coded “doctor” self-certainty, β = .52, t(87) = 2.86, p < .01. When entered simultaneously into the regression with support condition, greater coded “doctor” self-certainty predicted greater self-concept clarity (Figure 4), β = .64, t(87) = 7.75, p < .001. Assigned support condition did not remain a significant predictor of self-concept clarity in this analysis, β = .15, t(87) = 1.51, p = .13, and revealed a total indirect effect of condition with a 95% BCa bootstrap CI of 0.07 to 0.54. Zero falls outside this CI, indicating significant mediation at the p < .05 level.

Figure 4.

Study 2: Coded “doctor” self-certainty mediates the relationship between assigned support condition and self-concept clarity.

The results of Study 2 demonstrated that, after a threat to an important self-aspect, receiving evidentiary support from a friend increased individuals’ certainty that they possessed the specific self-aspect that had been threatened and increased individuals’ self-concept clarity. Although all forms of social interaction with a friend boosted mood to a similar degree, only evidentiary support increased self-concept clarity. Evidentiary support benefited certainty in the threatened self-aspect and self-concept clarity to a greater degree than either receiving emotional warmth from a friend or having a pleasant and distracting conversation with a friend. Study 2 also replicated Study 1’s evidence for a mediational pathway such that participants’ certainty that they possessed the specific threatened self-aspect significantly accounted for the relationship between receiving evidentiary support after a self-concept threat and increased self-concept clarity.

General Discussion

In the current research, we proposed a form of supportive social interaction, called evidentiary support, in which another person provides individuals with data and evidence for a specific threatened aspect of their self-concept. We hypothesized that evidentiary support would help repair both the specific threatened self-aspect and the overarching clarity of the self-concept after a threat, whereas emotional warmth would only assist in mood repair. Both studies supported our hypotheses. Whether through imagined (Study 1) or face-to-face interaction (Study 2), having a friend offer evidentiary support for a cherished threatened self-aspect was superior to receiving emotional warmth or a pleasant distraction in terms of repairing confidence in the self-aspect under threat and stabilizing overall self-concept clarity. In addition, participants’ certainty that they possessed the self-aspect that had been threatened mediated the relationship between receiving evidentiary support after a self-concept threat and increased self-concept clarity. Finally, the differential effects of evidentiary support on self-concept repair as opposed to the more general effects of social interaction on mood repair were demonstrated in Study 2.

Implications and Future Directions

The findings from the current research provide a novel, interaction-focused perspective on how individuals directly defend their self-concepts to maintain their existing views of who and what they are in the face of a threat. Specifically, the current research demonstrated that, when an important aspect of the self-concept has been threatened, receiving evidence for their threatened self-concept aids in confirming and stabilizing their sense of who they are. Taken together, these findings integrate social psychological perspectives on the self as both evidence-based and socially constructed to propose a specifically interaction-based means by which individuals directly maintain their self-concepts when faced with threatening circumstances.

The findings from the current research also have a variety of implications for how psychologists understand the process of self-maintenance and change. The current research adds to the literature within social psychology highlighting the importance of individuals’ interactions in the social world in the creation (e.g., Mead, 1934) and maintenance (e.g., Swann, 1983) of their self-concepts. Results also provide support for the importance of the hierarchical structure and perceived consistency of the self-concept. Specifically, one of the most novel aspects of the current research is its integration of the social and evidence-based nature of the self to show that social interactions that directly deal with a threat to the self-concept by recognizing specific, lower level (in terms of hierarchical structure) self-aspects can help to confirm individuals’ self-concept content and consistency better than social interactions that indirectly deal with a threat to the self-concept by recognizing the general, higher level (in terms of hierarchical structure) notion of the self as positive and valued. Indeed, the specific effects of evidentiary support on self-concept clarity in Study 2 was contrasted with the general and positive effects that all types of friendship support had on mood repair.

These results also carry implications for understanding when individuals may more successfully manage self-concept threats. Individuals are likely to experience self-threats in varied domains during their lives, but perhaps one determinant of how individuals’ self-concepts emerge from threatening circumstances is whether they have social others available to them who can provide them with evidentiary support. For example, when an individual transitions into an environment with a new social circle (e.g., a cross-country move, the transition into college), his or her new friends may be able to provide emotional warmth and acceptance and thus mood repair after a self-threat, but are unlikely to possess the deep knowledge of him or her that is necessary to provide evidentiary support. Without access to such support, individuals’ self-concepts may be more vulnerable to change and reductions in clarity after threats. Indeed, these types of life transitions are often highlighted as catalysts of self-change (Brammer, 1992; Erikson, 1968; Simmons, 1987), and it is possible a loss of evidentiary support may provide a partial explanation.

The findings from the current research also have broader implications for the role that social others play in maintaining individuals’ selves. Specifically, although the threats in the current research were failure-based threats to cherished aspects of most participants’ selves, the threats that individuals experienced in the lab were transitory. What would happen if the threat individuals faced provided irrefutable data that they did not possess a self-aspect they had previously considered to be part of their self-concept? Receiving evidentiary support for a threatened aspect of their self-concept may not be beneficial to individuals who are faced with a threat or failure that cannot be overcome easily, such as a failure to graduate from college. Indeed, receiving support indicating that they do possess a self-aspect that has been irreparably threatened may be counterproductive as it may keep individuals attached to, and believing that they possess, a self-aspect that can no longer realistically be considered true of them. Future research should investigate how the role that social others and evidentiary support may shift depending on the severity or permanence of the threat individuals experience to their self-concepts.

Similarly, the evidentiary support that social others provided individuals in the current research was either constrained to be or recalled as appropriate and applicable to the aspect of their self-concept that had been threatened. What would happen if this had not been the case? Future research should consider the consequences for the self-concept as well as the relationship when someone does not have the ability to provide individuals with evidentiary support or provides inappropriate evidentiary support.

Finally, just as differences may exist with regard to social others ability to provide individuals’ with evidentiary support for a threatened aspect of their self-concept, differences may exist with regard to individuals’ ability to respond to and take in the evidentiary support provided to them. The current research did not investigate whether individual differences exist in how responsive individuals are to receiving evidentiary support in response to a threat to their self-concept. Future research would benefit from investigating assorted possibilities for individual differences that might moderate the effects of the current research.

Limitations and Strengths

We wish to note several limitations of the current research. First, the current research only examines the effects of evidentiary support on individuals’ self-concepts at a single point in time and within a sample of undergraduate students. Existing research suggests that the stability of individuals’ self-concepts varies across the life span (e.g., Caspi & Roberts, 2001); thus, future research should examine whether evidentiary support functions similarly to confirm the threatened self-concepts of various age groups and across multiple time points. We would predict that evidentiary support functions similarly confirm a threatened self-concept over the life span; however, experiences being construed as threats to the self may become less frequent with age, while at the same time the amount of “evidence” close others know about your important self-aspects may increase with age (given that our close relationships to our spouses, siblings, and friends age along with us).

The present research also focused on threats to positive self-aspects, so the question of whether negative self-aspects would also be protected through this specific strategy remains untested. On the one hand, those closest to us are quite likely to possess evidence of our negative as well as our positive self-aspects, and robust evidence demonstrates that we prefer our closest others to know us, warts and all (e.g., De La Ronde & Swann, 1998; Kwang & Swann, 2010; Swann, Hixon, & De La Ronde, 1992). On the other hand, they may be less willing to share memories of our failings as readily as our strengths, making evidentiary support for negative self-aspects perhaps less likely to spontaneously occur. Future research should examine both the effectiveness and the prevalence of evidentiary support in maintaining both positive and negative self-aspects.

Finally, the current research contrasts receiving evidentiary support with receiving emotional warmth after a threat to the self to find that evidentiary support stabilizes self-concept structure and clarity better than emotional warmth. However, receiving support that helps one feel socially cared for (conveyed by emotional warmth) certainly has variety of benefits for individuals related to overall self-esteem. Although emotional warmth was not related to self-concept content and perceived consistency in the current research, it allows individuals to reflect on their social connections and so should provide benefits to individuals that are unrelated to self-concept content and consistency (e.g., Crocker et al., 2008). Certainly, the beneficial effects of emotional support on mood repair (even the absence of influencing self-concept repair) seen in Study 2 support this notion. Additional work from the social support literature has, in fact, identified specific esteem-focused types of support (e.g., Cohen, Mermelstein, Kamarck, & Hoberman, 1985). The current research is one of the first to focus on the specific interaction qualities that might constitute a self-concept structure or clarity focused type of support. Thus, research should investigate how emotional warmth might interface with evidentiary support to predict differential self-relevant outcomes, such as self-esteem, after a self-relevant threat. Especially in the context of close relationships, evidentiary support and emotional warmth may be highly likely to co-occur, and this co-occurrence may have interesting implications for individuals’ self-concepts.

We also wish to highlight several strengths of the current research. First, the current research uses varied methodologies. In terms of self-threat, the current set of three studies used naturalistic recall of past threats to the self-concept and direct experimental manipulations of success versus failure in self-relevant domains. In terms of supportive social interactions, the studies used both imagined and actual supportive interactions. In addition, the dependent measures used in the current research included self-reports of specific self-aspects and overall self-concept clarity and coded measures of expressions of both self-content and clarity during social interactions, providing convergent validity across the studies.

Second, the current research extends the investigation into how individuals engage in social means of maintaining their self-concepts and thus complements research within the domains of self-completion, self-verification, and self-affirmation. Specifically, the current research extended beyond the social displays of self-completion theory to investigate how social others might directly support individuals’ threatened self-concepts in the context of interactions. In addition, the current research extended the existing work on self-verification theory (e.g., De La Ronde & Swann, 1998; Swann, 1983; Swann & Predmore, 1985) by both focusing on one specific mechanism through which close others may help us maintain specific self-views, and by examining its influence on the metacognitive consequences of self-concept clarity in addition to protecting the specifically threatened self-view. The current research also extends beyond the contributions of self-affirmation theory (e.g., Steele, 1988). The results of the studies demonstrated that affirming the self-concept as positive and socially valued—as emotional warmth could be conceived of as support for the self as worthy of affection—was insufficient to repair individuals’ threatened self-concept content or perceived consistency, despite making individuals feel better. How, precisely, evidentiary support might further extend self-completion and self-verification theory, and interface with self-affirmation theory remains a fascinating question for future research.

Conclusions

Taken together, results of the current work provide evidence for a direct, social, distinctly interaction-based pathway through which individuals maintain a certain and consistent self-concept. Three studies demonstrated that when their sense of self had been threatened in some way, receiving evidence to confirm that they did, in fact, possess the threatened part of their identity helped individuals to maintain a certain and consistent view of who they are. Thus, the current research suggests that when a threat calls a cherished self-aspect into question, individuals turn to others to confirm and stabilize the self-concept, to “sing back the song in [their] hearts when [they] have forgotten the words.”

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Alicke

Sedikides

(2009). Self-enhancement and self-protection: What they are and what they do. European Review of Social Psychology, 20, 1-48.

Anderson

N. H.

(1968). Likableness ratings of 555 personality-trait words. Journal of Personality and Social Psychology, 9, 272-279.

Aron

(2003). Self and close relationships. In Leary

M. R.

Tangney

J. P.

(Eds.), Handbook of self and identity (pp. 442-461). New York, NY: Guilford Press.

Aronson

Blanton

Cooper

(1995). From dissonance to disidentification: Selectivity in the self-affirmation process. Journal of Personality and Social Psychology, 68, 986-996.

Baumeister

R. F.

(2010). The self. In Baumeister

R. F.

Finkel

E. J.

(Eds.), Advanced social psychology: The state of the science (pp. 139-176). London, England: Oxford University Press.

Baumeister

R. F.

Leary

M. R.

(1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117, 497-529.

Bem

D. J.

(1972). Self-perception theory. In Berkowitz

(Ed.), Advances in experimental social psychology (6th ed.). New York, NY: Academic.

Bigler

Neimeyer

G. J.

Brown

(2001). The divided self revisited: Effects of self-concept clarity and self-concept differentiation on psychological adjustment. Journal of Social & Clinical Psychology, 20, 396-415.

Brammer

L. M.

(1992). Coping with life transitions. International Journal for the Advancement of Counseling, 15, 239-253.

10.

Brown

J. D.

Smart

S. A.

(1991). The self and social conduct: Linking self-representations to prosocial conduct. Journal of Personality and Social Psychology, 60, 368-375.

11.

Campbell

J. D.

(1990). Self-esteem and the clarity of the self-concept. Journal of Personality and Social Psychology, 59, 538-549.

12.

Campbell

J. D.

Assanand

Paula

A. D.

(2003). The structure of the self-concept and its relations to psychological adjustment. Journal of Personality, 71, 115-140.

13.

Campbell

J. D.

Trapnell

P. D.

Heine

S. J.

Katz

I. M.

Lavallee

L. F.

Lehman

D. R.

(1996). Self-concept clarity: Measurement, personality correlates, and cultural boundaries. Journal of Personality and Social Psychology, 70, 141-156.

14.

Caspi

Roberts

B. W.

(2001). Personality development across the life course: The argument for change and continuity. Psychological Inquiry, 12, 49-66.

15.

Cohen

McKay

(1984). Social support, stress, and the buffering hypothesis: A theoretical analysis. In Baum

Taylor

S. E.

Singer

(Eds.) Handbook of psychology and health (Vol. 4, pp. 253-268). Hillsdale, NJ: Lawrence Erlbaum.

16.

Cohen

Mermelstein

Kamarck

Hoberman

H. M.

(1985). Measuring the functional components of social support. In Sarason

I. G.

Sarason

B. R.

(Eds.), Social support: Theory, research and applications (pp. 73-94). Amsterdam, The Netherlands: Springer.

17.

Cresci

M. K.

(2001). The relationship between informational support and post-injury functional status in older women recovering from a hip fracture. Educational Gerontology, 27, 281-295.

18.

Crocker

Niiya

Mischkowski

(2008). Why does writing about important values reduce defensiveness? Self-affirmation and the role of positive, other-directed feelings. Psychological Science, 19, 740-747.

19.

Cutrona

C. E.

(1990). Stress and social support: In search of optimal matching. Journal of Social & Clinical Psychology, 9, 3-14.

20.

Cutrona

C. E.

Russell

D. W.

(1990). Type of social support and specific stress: Toward a theory of optimal matching. In Sarason

B. R.

Sarason

I. G.

Pierce

G. R.

(Eds.), Social support: An interactional view (pp. 319-366). New York, NY: Wiley.

21.

Ronde

Swann

W. B.

Jr. (1998). Partner verification: Restoring shattered images of our intimates. Journal of Personality and Social Psychology, 75, 374-382.

22.

Epstein

(1973). The self-concept revisited: Or a theory of a theory. American Psychologist, 28, 404-416.

23.

Erikson

E. H.

(1968). Identity: Youth and Crisis. NewYork, NY: Norton.

24.

Festinger

(1957). A theory of cognitive dissonance. Stanford, CA: Stanford University Press

25.

Goffman

(1959). The presentation of self in everyday life. Oxford, UK: Doubleday.

26.

Gollwitzer

P. M.

Wicklund

R. A.

(1985). Self-symbolizing and the neglect of others’ perspectives. Journal of Personality and Social Psychology, 48, 702-715.

27.

Greenberg

Pyszczynski

(1985). Compensatory self-inflation: A response to the threat to self-regard of public failure. Journal of Experimental Social Psychology, 49, 351-372.

28.

Harlow

R. E.

Cantor

(1995). To whom do people turn when things go poorly? Task orientation and functional social contacts. Journal of Personality and Social Psychology, 69, 329-340.

29.

House

(1981). Work stress and social support. Reading, MA: Addison-Wesley.

30.

James

(1890). The principles of psychology. Cambridge, MA: Harvard University Press.

31.

Kwang

Swann

W. B.

Jr. (2010). Do people embrace praise even when they feel unworthy? A review of critical tests of self enhancement versus self-verification. Personality and Social Psychology Review, 14, 263-280.

32.

Leary

M. R.

Baumeister

R. F.

(2000). The nature and function of self-esteem: Sociometer theory. In Zanna

(Ed.), Advances in experimental social psychology (Vol. 32, pp. 1-62). San Diego, CA: Academic Press.

33.

Malecki

C. K.

Demeray

M. K.

(2003). What type of support do they need? Investigating student adjustment as related to emotional, informational, appraisal, and instrumental support. School Psychology Quarterly, 18, 231-252.

34.

Markus

(1977). Self-schemata and processing information about the self. Journal of Personality and Social Psychology, 35, 63-78.

35.

Markus

Wurf

(1987). The dynamic self-concept: Social psychological perspective. Annual Reviews of Psychology, 38, 299-337.

36.

McConnell

A. R.

(2011). The multiple self-aspects framework: Self-concept representation, and its implications. Personality and Social Psychology Review, 15, 3-27.

37.

Mead

G. H.

(1934). Mind, self, and society, Chicago, IL: University of Chicago Press.

38.

Nezlek

J. B.

Plesko

R. M.

(2001). Day-to-day relationships among self-concept clarity, self-esteem, daily events, and mood. Personality and Social Psychology Bulletin, 27, 201-211.

39.

Pickett

C. L.

Gardner

W. L.

(2005). The social monitoring system: Enhanced sensitivity to social cues as an adaptive response to social exclusion. In Williams

Forgas

Von Hippel

(Eds.), The social outcast: Ostracism, social exclusion, rejection, and bullying (pp. 213-226). New York, NY: Psychology Press.

40.

Preacher

K. J.

Hayes

A. F.

(2004). SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behavior Research Methods, Instruments, & Computers, 36, 717-731.

41.

Ritchie

T. D.

Sedikides

Wildschut

Arndt

Gidron

(2011). Self-concept clarity mediates the relation between stress and subjective well-being. Self and Identity, 10, 493-508.

42.

Sedikides

Gregg

A. P.

(2008). Self-enhancement: Food for thought. Perspectives on Psychological Science, 3, 102-116.

43.

Simmons

(1987). Social transition and adolescent development. New Directions in Child Development, 37, 33-61.

44.

Slotter

E. B.

Gardner

W. L.

Finkel

E. J.

(2010). Who am “I” without “you”? The influence of romantic breakup on self-concept clarity. Personality and Social Psychology Bulletin, 36, 147-160.

45.

Steele

C. M.

(1988). The psychology of self-affirmation: Sustaining the integrity of the self. In Berkowitz

(Ed.), Advances in experimental social psychology (Vol. 21, pp. 261-302). New York, NY: Academic Press.

46.

Swann

W. B.

(1983). Self-verification: Bringing social reality into harmony with the self. In Suls

Greenwald

A. G.

(Eds.), Psychological perspectives on the self (Vol. 2, pp. 33-66). Hillsdale, NJ: Lawrence Erlbaum.

47.

Swann

W. B.

Jr. Hill

C. A.

(1982). When our identities are mistaken: Reaffirming self-conceptions through social interaction. Journal of Personality and Social Psychology, 43, 59-66.

48.

Swann

W. B.

Jr. Hixon

J. G.

Ronde

(1992). Embracing the bitter truth: Negative self-concepts and marital commitment. Psychological Science, 3, 118-121.

49.

Swann

W. B.

Jr. Predmore

S. C.

(1985). Intimates as agents of social support: Sources of consolation or despair? Journal of Personality and Social Psychology, 49, 1609-1617.

50.

Taylor

S. E.

(2007). Social support. In Friedman

H. S.

Silver

R. C.

(Eds.), Foundations of health psychology (pp. 145-171). New York, NY: Oxford University Press.

51.

Tesser

(2000). On the confluence of self-esteem maintenance mechanisms. Personality and Social Psychology Review, 4, 290-299.

52.

Tice

D. M.

Baumeister

R. F.

(2001). The primacy of the interpersonal self. In Sedikides

Brewer

M. B.

(Eds.), Individual self, relational self, collective self (pp. 71-88). Philadelphia, PA: Psychology Press.

53.

Watson

Clark

Tellegen

(1988). Development and validation of brief measures of positive and negative affect: The PANAS Scales. Journal of Personality and Social Psychology, 54, 1063-1070.

54.

Wicklund

R. A.

Gollwitzer

P. M.

(1982). Symbolic self-completion. Hillsdale, NJ: Lawrence Erlbaum.